*The Catalyst: RNA and the Quest to Unlock Life’s Deepest Secrets, Thomas R. Cech, 2024. Cech won the Nobel Prize in Chemistry in 1989 (with Sidney Altman) for the discovery of the catalytic properties of RNA, which means that in addition to carrying genetic information, RNA could also act on other molecules.

Reading this with the inaccurately-named “26-minute book club” in the Spring of 2025.

Contents

My Take on the Book

A worthwhile read! I learned a huge amount about RNA that I didn’t know — not surprising since my biochemistry education ceased in the late 1970’s. RNA is not just about mechanisms for producing proteins — though I will pause and note that its roles in doing that are much more complex than I’d understood — but it does everything from encoding information to acting a a biological catalyst. Here’s the running list I kept of types of RNA (though no doubt a few were overlooked):
mRNA, tRNA, rRNA, snRNA, ribozyme, RNAse P, siRNA, microRNA, telomerase

The drawbacks I find in the book are really those of style, stemming from a (laudable) effort to reach a general audience. It is a bit wordy, and uses what I see as rather clumsy analogies to help general readers understand what’s going on. There is also a bit too much emphasis on the work of different scientists and labs, and their cooperation/competition… most likely to add a bit of drama to the book. But one can skim those bits, and it has plenty of good science in it.

Types:

The Book

Introduction: The Age of RNA

While public attention has been focused on DNA, we have been gradually learning more and more about RNA. While DNA is essentially a one-trick pony, a molecule that can encode genetic information and store it in a stable form, it turns out that variants of RNA can do many different things.

The first section of the book discusses what we have learned about RNA.

It constructs proteins
It mediates gene splicing, enabling complex organisms like humans to get far more out of their genomes that simpler organisms.
It can function like enzymes, in which case it is called a ribozyme
It works in tandem with ribosomes to read mRNA and construct proteins

The section ends with a discussion of how RNA may hold the secret to the emergence of life.

The second section focuses on applications, and ranges from telomerase to mRNA vaccines and CRISPR.

PART I: THE SEARCH

C1. The Messenger (mRNA & tRNA)

1944: Oswald Avery discovers that DNA is the molecule that encodes genetic information.
1953: Crick and Watson discover the structure of the genetic code.
1953: Gamow‘s quest to use math and physics to decipher the genetic code. They focused on RNA, because DNA is found only in the nucleus, but RNA is found everywhere
1954: Gamow forms the RNA tie club, with each of twenty scientists assigned to figure out the code of a particular amino acid. But by the late 50’s the club had solved nothing, and Gamow appears to have gone on to other things.
A puzzle: The proportion of bases in the RNA of a cell is quite stable over time. Also, most of the cell’s RNA appears to be long lived. This seems inconsistent with the fact that cells can switch between the synthesis of different proteins in a matter of seconds or 10s of seconds (either as they respond to different condition, or when they are infected by phages).
Discovery/Characterization of mRNA. Responding to another experiment which showed that a phage-infected bacterium started producing RNA that had a lower molecular weight, Brenner, Jacob, and Meselson did an experiment where they infect a bacterium with a phage, and radioactive uracil, and then use an ultracentrifuge to separate RNA by size. They demonstrated that the new (m-)RNA was produced, and incorporated into the ribosomes.
((Some stats on how quantities and speeds within cells:
(1) a bacterial cell contains 15-20K ribosomes;
(2) Each ribosome can add 15-20 amino acids/second to a protein;
(3) a cell can produce 4.5K proteins per second. ))
mRNA was difficult to detect because it comprises only about 5% of the RNA in a cell, and furthermore, the composition of mRNA differs depending on which protein it produces, and eColi have about 4,000 different proteins (and therefor mRNAs) Also mRNAs have lifespans in minutes, whereas ribosomal RNA is very long-lived.
The human genome has about 3 billion bases or letters, packed into 13 chromosomes.
Codons: groups of 3 bases. AUG == start; UAG == stop. After a stop codon each group of three bases codes for an amino acid until a stop codon is reached.
Crick et al. proved that codons consisted of three bases by mutating a phage gene by inserting molecules of an acrimide dye: one and two insertions result in nonsense; but 3 insertions lead to the generation of (mostly) correct amino acids.
Deciphering the correspondence between RNA codons and Amion Acids.
- Nirinberg and Matthaei discovered that one can create mRNA’s composed of a single base (i.e. UUU), and that in turn leads to the production of a protein consisting of a single amino acid. Thus UUU, AAA, and CCC can be decoded.
- Khorana discovered how to synthesize particular DNA sequences, and these, translated to RNA, enabled them to deduce the amino acids that were coded for.
Discovering tRNA used to actually construct proteins from mRNA
- Crick deduces that there must be a molecule where one end will bind to an RNA codon, and the other end will bind to an amino acid.
- Zamecnik demonstrates existence of tRNA.

C2. Splice of Life (snRNA marks ends of introns)

Introns. Discovery that some regions of nuclear RNA are not copied to mRNA. Instead different parts are spliced together into an mRNA.
From one, many: It turns out that splicing enables an organism to create multiple proteins from the same DNA sequence. This explains why much more complex organisms like humans do NOT have proportionally more complex genomes.
snRNA: Small nuclear RNAs mark the ends of introns, allowing the regulation of splicing.

C3. Going It Alone (ribozyme & RNAse P)

Misapprehension 1: What had been a fundamental dogma since the 1920’s — “all enzymes are proteins” — turns out to have been wrong.
Misapprehension 2: Initially, most scientists thought of RNA simply as an intermediary, supporting the creation of proteins based on a segment of a DNA nucleotide sequence.
Tetrahymena thermophilia. As often seems to be the case, progress came from selecting an organism with unusual characteristics that gives scientists pursuing particular questions special traction. In this case it was tetrahymena thermophilia, a form of ‘pond scum’ that has the unusual characteristic of having large number (10,000 rather than 2) of easily isolable genes (not part of the chromosomal structure) for its ribosomal RNA. This makes it possible to isolate the genes for ribosomal RNA.
Intron excision and RNA splicing found to occur. “Our goal was to understand how these Tetrahymena genes were transcribed into RNA and how the proteins bound to the DNA— a special feature of eukaryotic chromosomes-might regulate this process.”
- As it turned out, the particular gene being examined had an intron, and that they could observe the excision of the intron and thus the splicing of the final tRNA gene in the test-tube. After various experiments to try to isolate the presumed enzyme that was doing the excision/splicing, they discovered that it would still occur even if they had eliminated all proteins.
- At around the same time, Paula Grabowski isolated one of the excised introns and discovered that it would convert to a circular form, even in the absence of a protein-enzyme.
RNA as a self-catalyst. These two observations, and a subsequent experiment that produced the RNA gene from eColi rather than tetrahymena, provided weight to the hypothesis that a form RNA was serving a catalytic role without requiring the presence of a protein.
RNA as an external catalyst. Not long after, Sid Altman, who had been working on RNAse P, discovered that it could catalyze reactions involving external RNA.
Summary: RNA can do a lot. “In our experiments, we had discovered an RNA that could splice itself-that could be its own internal catalyst. Now Sid’s team had found that RNA could work as an external catalyst, acting upon something else— tRNA precursors. In both cases, RNA had emerged as a molecule that was not merely carrying information from DNA to protein but was an active driver of cellular reactions.”
AND snRNAs too. And yet a bit later, scientists in Australia discovered that snRNAs can also catalyze reactions.

…reading break…

Nobel Prize winners for work related to genetics

1962: Watson, Crick, Wilkins
Discoveries concerning the molecular structure of nucleic acids and its significance for information transfer in living material.

1965: Jacob, Lwoff, Monod
Discoveries concerning genetic control of enzyme and virus synthesis. [Also discovered RNA, but prize not awarded for that]

1968: Holley, Khorana, Nirenberg
Interpretation of the genetic code and its function in protein synthesis

1975: Baltimore, Dulbecco, Temin
Discoveries concerning the interaction between tumour viruses and the genetic material of the cell

1978: Arber, Nathans, Smith
Discovery of restriction enzymes and their application to problems of molecular genetics.

1980: Berg, Gilbert, Sanger
Fundamental studies of the biochemistry of nucleic acids, with particular regard to recombinant-DNA and contributions concerning the determination of base sequences in nucleic acids.

1983: McClintock
Discovery of mobile genetic elements

1989: Altman, Cech
Discovery of catalytic properties Of RNA, discovering that RNA actively aids chemical reactions in cells and can function as enzymes (ribozymes)

1993: Roberts, Sharp
Discovery of split genes

2024: Ambros, Ruvkun
Discovery of microRNA and its role in post-transcriptional gene regulation Nobelprize

2006: Fire, Mello
Discovery of RNA interference (RNAi), where specific mRNA molecules are inactivated by adding double-stranded RNA to cells.

2009: Ramakrishnan, Steitz, Yonath
Studies of the structure and function of the ribosome: mapping the ribosome at the atomic level, showing how it translates messenger RNA into protein

2020: Doudna, Charpentier
For the development of a method of genome editing. While CRISPR-Cas9 is primarily known as a DNA editing tool, the system is fundamentally guided by RNA molecules (tracrRNA and crRNA) that direct the Cas9 protein to specific DNA sequences for cutting.

2023: Karikó, Weissman
Discoveries concerning nucleoside base modifications that enabled the development of effective mRNA vaccines against COVID-19

C4: The Shape of a Shapeshifter (tRNA)

The last chapter showed that various forms of RNA could catalyze cutting and splicing of RNA, but did not provide insight on how RNA could do that. To understand this, scientists need to understand the 3D structures.

RNA has many shapes. RNA, being single-stranded, is not constrained to a helix like DNA, but can take on a vast number of shapes.
Keller determines 3-leaf clover (roadkill) shape of tRNA. In the 1950’s and 60’s, Bob Holly’s lab focused on tRNA from yeast that transferred alanine and isolated it. Elizabeth Keller, working in his lab, used pipe cleaners to figure out a 3-leaf clover structure, she and Holly reasoning that the anti-codon had to be exposed to support its function. Because all the tRNA’s had to fit into the same slot in the ribosome for protein synthesis to occur, it could be assumed that they would all have a similar structure.
3D structure of tRNA. In the late 1960’s Kim and Rich at MIT, and Robertus and Klug at Cambridge England both succeeded in characterizing the 3-D form of tRNA, showing that 3-leaf clover structure was correct, and describing the 3D shape it folded into.
Ribosomal RNA. After that attention turned to the 3D structure of ribosomal RNA: it took a long time but the components of ribosomes were gradually solved.
Crowdsource and AI to find 3D structure. After that the field progressed slowly but steadily. However, in the ‘aughts and later, progress has come on another front: first crowd-sourced contests to deduce 3D structures, and now the use of AI to determine structure. While lab work is still needed, it may well be the case that one day new 3D structures will be deduced solely by computers.
Synthetic Biology. Mention of ribozymes as sensors and switches in molecular circuits in an area known as synthetic biology.

C5: The Mothership (rRNA: small/ large subunits)

The Ribosome! Harry Noller’s investigation of Ribosomes.
How to make progress in sequencing by using analogous RNA from different organisms. Noller comes across work by Wose and Fox (1975) that characterized the structure of the smallest of the ribosomal RNAs by comparing sequences of that RNA from multiple organisms and assuming they all needed to fold up in the same way, Noller realizes he can take the same approach to the larger components of the ribosome.
Large Subunit Sequence. In the late 70’s Wose and Noller succeed in sequencing the two larger ribosomal RNA components.
Figuring out how to crystalize ribosomal RNA. Ada Yonath succeeds in crystalizing ribosomal RNA, but the crystals are not stable in the high salt solutions needed for xRay defraction, but she found success by shifting to ribosomal RNA from a Dead Sea bacterium that she reasoned would be stable in high salt environments. This worked out, but it took another decade to create the heavy metal version or ribosomal RNA needed for crystallographic characterization.
Functions of the two subunits. The small subunit of ribosomal RNA is the first to assemble with mRNA and is responsible for reading out the code in the mRNA and lining up the tRNAs. The large subunit contains the catalytic center that stitches together the amino acids.
3D structure of the large subunit. Tom Streitz succeeded in characterizing the 3D structure of the large subunit by inserting a group of 18 tungsten atoms into the subunit. When they did this they could see it was composed exclusively of RNA.
3D structure of the small subunit. Ramakrishna did the same for the small subunit. Almost all of it is RNA.
3D structure of functional complex. Next Noller and Cate succeeded in crystalizing the entire ribosome inhabited by its functional members — mRNA and tRNA. By comparing the x-ray of this crystal with one of Ramakrishnan’s images of an empty small subunit, they were able to deduce the structure of the functional ribosome. Noller, Streitz and Yonath receive the Nobel in 2009.
Practical applications — antibiotics. Human and bacterial ribosomes are sufficiently different that we can find substances that disrupt bacterial ribosomes but not human ones. This permits the development of antibiotics — about half of our antibiotics target bacterial ribosomes.

C6: Origins (RNA World)

The oldest life. Bill Schopf, a paleontologist, has found micro-fossils from 3.3 to 3.5 Ga.
The RNA World. The crux of the idea is that since RNA can act both as a carrier of information, but also as an ‘RNAzyme,’ it can serve as a self-replicating information carrier, and is thus a better candidate for the origin of life than the simultaneous appearance of DNA and the proteins necessary to replicate it. RNA can also be double-stranded, and can separate into single strands.
Spontaneous formation of precursors. Nucleic acids can form spontaneously. Similarly, very high concentrations of nucleic acids can form fragments of RNA.
in 1981 Szostak and Doudna re-engineered tetrahymena to be able to produce a complementary RNA sequence when given a template RNA — this includes the case where the template is its own RNA sequence. However they couldn’t do that for the full sequence. What they were able to do was to produced three shorter sequences that once produced would self-assemble into the full RNA.
Szostak’s group has produced protocells that nucleic acids within protocells can form longer nucleic acid sequences, and that such cells can grow and divide, but not with the regularity that living cells do. [What does ‘divide’ mean in this context?]

…reading break…

PART II: THE CURE

C7: Is the Fountain of Youth a Death Trap? (telomerase)

About the discovery and functions of telomeres, and telomerase.

Telomeres. DNA has repeating sequences on its ends called telomeres. As it replicates the telomeres are either maintained (by an enzyme called telomerase) or (in its absence) telomeres grow shorter and shorter with each replication. When the telomeres are gone the cell enters senescence — it will no longer divide and will eventually die.
The Hayflick limit. The number of times a cell can divide is called the Hayflick limit.
Ever-multiplying cells that maintain telomerase. Some types of cells — for example free-living cells like yeast and pond scum (e.g. tetrahymena) and stem cells in animals – have telemeroase and are able to keep replicating indefinitely.
Stem cell division. When a stem cell divides it produces one stem cell and a normal cell lacking telemerase.
Cells with Hayflick limits that have lost their telomerase. But most cells in multicellular organism lose their telemeroase as the develop, because the organism can not manage its structure with a bunch of infinitely replicating immortal cells — these are figuratively and literally a cancer.
Telomerase is RNA, though it requires a protein as an adjunct.
Oxyyticha nova, another form of pond scum, is an organism that has one gene per chromosome, and thus has 100,000 chromosomes — meaning that it has LOTS of telomeres.
Promoter sequence mutation turns telemerase back on.

C8: As the Worm Turns (RNAi: microRNA & siRNA)

The discovery of RNA interference (RNAi). Dicer binds to double stranded RNA and chops it into siRNA (small interfering RNA). The siRNA binds to a protein call Argonaute which jettisons one strand (called the passenger strand) leaving a single strand called the guide strand which binds to a particular mRNA which is then cleaved (and destroyed) by Argonaute.

microRNAs. It turns out that there are lots of very small microRNAs in cells, and these are the grist for RNAi.
microRNA targets. Each microRNA targets not a single mRNA but a class of related mRNAs.
There are at least 500 microRNAs in humans, coded for in the genome (only 2% of the genome codes for proteins).
Because this microRNAs target particular (related) mRNAs, and because microRNAs are small, they are good candidates for targeting single-mutation diseases, especially rare (“orphan”) diseases. May have promise for neurodegenerative diseases.

C9: Precise Parasites, Sloppy Copies

Viruses and RNA. Viruses come in two forms: those that store their information in DNA, and those that store it in RNA. RNA viruses can be positive or negative.
Positive RNA viruses produce a strand of negative RNA that serves as a template to replicate the infectious positive strand which then takes over the ribosomal machinery in the host cell.
Negative RNA infiltrates cells as the template, and with the help of a copying enzyme produces positive copies of the infectious strand and again takes over the ribosomal machinery.
Viruses need
(1) a way to enter a cell,
(2) a way to manufacture the proteins they need,
(3) a way to replicate their genomes, and
(4) a way to package infectious offspring (an envelope of capsid)
Efficiency. Viruses only code for a few (4 -29) proteins.
Mutations. Almost every replication of a virus results in one or more errors — this, along with their fast replication rate – is why viruses are so mutable.

…reading break…

RNA versus RNA

About Genetic Therapies. Gene therapy is the idea of making a change to a gene in the genome, and thereby permanently altering it for therapeutic purposes. A different idea was DNA therapy, where a curative version of DNA could be introduced, and would temporarily produce proteins that the immune system could learn to recognize and attack. A problem with DNA therapy, however, was that there was no way to control where the introduced DNA would end up, and it would sometimes end up in a place where it had unfortunate effects. Bob Malone and Inder Verma, at the Salk Insititute, circa 1989, saw mRNA was a way to avoid this… It took 30 years, but this is what lead to mRNA vaccines. The realization of mRNA vaccines required several different problems to be solved.
Mass Production of mRNA. Bill Studier at Brookhaven National Lab was investigating T7, a bacteriophage that was extraordinarily efficient at taking over bacterial cells. It produced an RNA polymerase that looked for genes with a particular 17 base-pair initiation site (which the phage genes had), and produced mRNAs that coded for phage proteins. In 1981 Studier realized that the phage RNA polymerase could be repurposed to produce any particular mRNA by adding the initiation sequence to the gene for that protein. He quickly found a way to mass produce phage RNA polymerase.
Delivering the mRNA into cells. Phil Felgner, at Syntax corporation, figured out how to use positively-charged lipid-walled ‘containers’ (liposomes) that could hold nucleic acids inside them. However, in 1988 his project was canceled, because it seemed to futuristic.
Reprogramming cells. Felgner moved to San Diego to work for Vical Corporation, and there was approached by Malone and Verma to collaborate on the delivery of mRNA into cells. The used mRNA for firefly luciferase — because it glowed and was therefore easy to detect — and soon found that Felgner’s liposomes were a successful delivery mechanism.
An alternate delivery mechanism. While Felgner’s liposomes worked well in cell cultures, they did not work so well in live animals — they caused drops in white cell production, blood clotting and inflammation, possibly because of the positively charged lipid membranes, which do not occur in nature. A different researcher, Pieter Cullis of UBC produced a lipid membrane that was neutral in a neutral environment, but that became positively charged in an acidic environment, such as is found inside a cell. Cullis’s vehicle could thus circulate through the blood stream and pass into the cell, where it will rupture (due to the charge) and release the mRNA. This vehicle was used as a delivery mechanism for mRNA cleaving siRNAs as a treatment for ATTR in a 2018 clinical trial.
Cloaking the mRNA. Another problem is that most organisms have an innate immunity to viral mRNA, so the mRNA needs to be disguised. Kariko and Wiseman developed a method for disguising RNA, replacing its “U” (uracil), with a modified version the called pseudo-U. PseudoU: (1) evaded innate immunity, (2) also evaded many RNA polymerases (making it more stable), but (3) still worked with Studier’s T7 phage polymerase. This was achieved circa 1998, and the technology licensed to a number of companies.
Covid sequenced, and the sequence released . On January 10, 2020, Prof. Yong-Zehn Zang of Fudan U released the sequence of the CoVID-10 virus on an open access site.
BioNTech and Moderna develop mRNA vaccines. Two then-obscure companies working on mRNA vaccines shifted their work to a CoVid vaccine. A problem was that the COVID virus’ spike flipped into a different form when it merged with a human cell, and so the immune system might be trained to look for the wrong shape. This was solved by altering the sequence for the spike protein to contain prolines, two very inflexible amino acids, in an attempt to stabilize the spike so that the immune system would be properly stimulated. (This technique had been developed for the MERs mRNA vaccine developed circa 2012 or later.)
Vaccine logistics. BioNTech partnered with Pfizer to address issues like clinical trials, vaccine storage and transport, and so on.
mRNAs as vaccines of the future? Most therapies work by inhibiting proteins involved in a disease producing process; but mRNA works by producing functional proteins that are mutated or absence in a disease situation. There are various possibilities. In theory, mRNAs for specific antibodies could be used to create immunity to a disease — but we don’t yet know that if mRNA therapies can deliver enough antibodies. Also, in the case where disease is caused by a missing or damaged protein, mRNA can be used to provide a functional form of that protein…

Running with Scissors (CRISPR)

The CRISPR story is told.
CAS9/CRISPR includes a guideRNA that binds to a particular DNA sequence, and tracrRNA that binds to the guideRNA and CAS9 so that CAS9 cleaves the DNA sequence at the right point.
NHEJ. When DNA in the cell is broken, there is a quick-and-dirty form of repair called Non-Homologous End Joining. This fuses broken pieces together, but does not look for any similarity in sequence, and often loses or inserts a few nucleotides.
Homologous recombination. Here the repair process looks for a matching DNA sequence to serve as a model for the repair, such as the other matching chromosome in a cell. However, for purposes of gene editing, researchers have found that they can provide a donor template that matches sequences near the break, but the also includes novel sequences which can be inserted during the repair.
Dead Cas9. Researchers have figured out how to inactivate Cas9 so that it does not cleave DNA, but so that it still binds to a specific place. And they can attach other proteins — those that activate and repress certain genes — so that they are delivered to a precise location. Another thing that can be done (David Liu) is to attach what are called base-editing proteins (which will change one nucleotide to another) to Dead Cas9, and erase single-base mutations, which are responsible for a variety of diseases.

RNA is up to its same old tricks. In every iteration of CRISPR, a guide RNA uses the power of nucleic acid base-pairing to bring the editing machinery to a specific site in a complex genome. Then an associated protein —whether it be catalytically active Cas9, Dead Cas9, or some other member of the family— performs some action on that DNA sequence.

We’ve encountered this principle before, in both RNA interference and telomerase: RNA provides the guide, and, once the RNA finds the site of action, a protein enzyme performs a catalytic act. Part of the reason why CRISPR gene-editing exploded so quickly is that it conforms to the pattern established by these earlier RNA breakthroughs. Scientists were given a revolutionary new pair of scissors, but they already knew how to run with them.

— ibid,, p. 229

Epilogue: The Future of RNA

Coding, Introns, Junk. The coding regions of human genes make up only about 2% of our genome. When we add the introns that interrupt the coding sequences and add flexibility, we get up to 24%. The other 75% was a mystery, until recently.
lncRNA. The 75% of our genome that consists of ‘dark matter’ DNA is transcribed into RNA referred to as long noncoding RNA (lncRNA). While the function of a few lncRNAs have been determined, the functionality of most lncRNAs is unknown.
Synthetic RNAs, e.g. aptamers. It is possible to generate vast numbers of RNA sequences that fold up into different forms, and filter them for specific versions that bind to particular molecules. These can be used as bio-sensors, and as ways of inactivating particular proteins or other molecules. There are a few companies, and clinically approved treatments, that use this approach.

# # #